Research map / 3D vision / Depth and stereo
Dense depth prediction and completion
Predicts consistent, high-resolution or sparse-input-completed depth from single images and video.
85 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Depth and stereo, keyword cluster “depth estimation · monocular · depth map”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
16 papers in 2023–24 and 13 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 5.2% to 3.2%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Single-Image Depth Perception in the Wild (NeurIPS 2016 · 97 citations)
- 2017 · Semantic Scene Completion From a Single Depth Image (CVPR 2017 · 1,301 citations)
- 2018 · Deep Ordinal Regression Network for Monocular Depth Estimation (CVPR 2018 · 1,971 citations)
- 2019 · DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene From Sparse LiDAR Data and Single Color Image (CVPR 2019 · 420 citations)
- 2020 · 3D Photography Using Context-Aware Layered Depth Inpainting (CVPR 2020 · 287 citations)
- 2021 · NeuralRecon: Real-Time Coherent 3D Reconstruction From Monocular Video (CVPR 2021 · 313 citations)
- 2022 · Neural Window Fully-Connected CRFs for Monocular Depth Estimation (CVPR 2022 · 358 citations)
- 2023 · DesNet: Decomposed Scale-Consistent Network for Unsupervised Depth Completion (AAAI 2023 · 28 citations)
- 2024 · Depth Information Assisted Collaborative Mutual Promotion Network for Single Image Dehazing (CVPR 2024 · 165 citations)
- 2025 · DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation (CVPR 2025 · 47 citations)
- 2026 · Depth-Synergized Mamba Meets Memory Experts for All-Day Image Reflection Separation (AAAI 2026 · 1 citation)
Earliest papers here
- Single-Image Depth Perception in the Wild (NeurIPS 2016 · 97 citations)
- Depth from a Single Image by Harmonizing Overcomplete Local Network Predictions (NeurIPS 2016 · 39 citations)
- Semantic Scene Completion From a Single Depth Image (CVPR 2017 · 1,301 citations)
What has been tried
73 papers here report 729 measured changes (a component added, removed or swapped, compared with a baseline): helped 75%, no clear effect 4%, hurt 20%, mixed 1%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add LEGM | SOTS-indoor · PSNR | 35.69 -> 40.41 | Helped | Depth Information Assisted Collaborative Mutual Promotion Network for Single Image Dehazing (CVPR 2024 · 165 citations) |
| Use bilateral preprocessing before multi-modal fusion | NYUv2 · RMSE (mm) | 91.68 -> 89.56 | Helped | Bilateral Propagation Network for Depth Completion (CVPR 2024 · 50 citations) |
| add depth prior to self-attention | NYU DepthV2 · mIoU | 51.7 -> 54.3 | Helped | DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation (CVPR 2025 · 47 citations) |
| add densification and a degradation bridge, and concatenate RGB and degradation features | DIML test split · RMSE (mm) | 108 -> 95 | Helped | Completion as Enhancement: A Degradation-Aware Selective Image Guided Network for Depth Completion (CVPR 2025 · 11 citations) |
| add first-stage pre-training | KITTI · RMSE | 682.34 -> 678.12 | Helped | Distilling Monocular Foundation Model for Fine-grained Depth Completion (CVPR 2025 · 10 citations) |
| Exclude novel views during reconstruction | Replica room-2 · Depth L1 | 1.53 -> 1.80 | Hurt | IBD-SLAM: Learning Image-Based Depth Fusion for Generalizable SLAM (CVPR 2024 · 6 citations) |
| Add attention layers for information sharing between views | KITTI-360 · Oacc | -1.67% | Hurt | Boosting Self-Supervision for Single-View Scene Completion via Knowledge Distillation (CVPR 2024 · 3 citations) |
| add feature fusion (Eq.3) | KITTI DC 1-shot · RMSE | 2.17±0.99 -> 1.68±0.07 | Helped | A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · 3 citations) |
| Add mesh constraints | UDIS-D · Visual distortion | Helped | Depth-Supervised Fusion Network for Seamless-Free Image Stitching (NeurIPS 2025 · 0 citations) | |
| add cross-view consistency loss | NYUv2 + video frames · RMSE | 0.4879 -> 0.4850 | Helped | Multi-task Learning with 3D-Aware Regularization (ICLR 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 5 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- SUN RGBD · 2 of 5 papers
- NYU DepthV2 · 2 of 5 papers
Baselines
- S2D · 2 of 5 papers
- CSPN · 2 of 5 papers
- TPVD · 2 of 5 papers
- GuideNet · 2 of 5 papers
- NLSPN · 2 of 5 papers
- CFormer · 2 of 5 papers
- RigNet · 2 of 5 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 29 papers; from OpenAlex and the papers' affiliations).
- Nanjing University of Science and Technology · 2 papers
- National University of Defense Technology · 2 papers
- KU Leuven · 2 papers
- ETH Zurich · 2 papers
Use it
Download BibTeX of 39 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Self-supervised monocular depth (116 papers)
- Foundation models for monocular geometry (81 papers)
- Depth from light field and defocus (60 papers)
Elsewhere in the topic
- Feed-forward 3D reconstruction (133 papers)
- Scene coordinate regression for localization (112 papers)
- Learned optical and scene flow (108 papers)
- Neural light field and illumination imaging (107 papers)
- Cost-volume multi-view stereo (107 papers)
- Camera model and calibration for SfM (99 papers)
- Non-line-of-sight imaging (93 papers)
- Dense feature matching for correspondence (80 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).